Q&A: Cesar Victora – 30 years of Brazil cohort studies
Birth cohort studies follow large numbers of people from birth, collecting information on the health, environmental and socio-economic factors that may affect their long-term health outcomes and human capital achievements. The Wellcome Trust funds three cohort studies in Pelotas, a city in the south of Brazil, run by Cesar Victora, Professor of Epidemiology at the Federal University of Pelotas.
Cesar was influential in setting up the Consortium of Health Orientated Research in Transitioning Societies (COHORTS), a collaboration between five large cohort studies in low- or middle-income countries. This initiative is showing the importance of nutrition in early life. Benjamin Thompson spoke to Cesar on his recent visit to London.
You’re an epidemiologist, but I understand you started off as a medical doctor?
Yes I started as an MD. I worked in a slum and saw the same kids coming back all the time with the same health problems, some of whom would die.
What sort of diseases are we talking about?
Malnutrition, diarrhoea, pneumonia, measles: common diseases of childhood and all ultimately preventable. I felt that I had to do something better than just treat the problems. That’s when I decided to do epidemiology.
So you moved over to England?
Yes, I came to the London School of Hygiene and Tropical Medicine in 1980 to study for a PhD. A colleague of mine called Fernando Barros from Pelotas was also studying here. He went back to Brazil earlier than I did, and we began the cohort. He was actually the one who was in charge of the study day-to-day in 1982, as I didn’t return until 1983. When I came back I got some funding to continue the study and here we are now 30 years later.
Why did the Pelotas cohort study begin?
It all began in the early 80’s when we didn’t have enough data on low birthweight, preterm birth and infant mortality in Brazil, because the statistics were not very good. We decided to do a population based study, in which we included all the births in our home city during a one year period – 1982. At that stage we thought it would only be a short study on child health. We never thought that 30 years later we would still be following up!
Why did you pick Pelotas?
We didn’t pick Pelotas for any special reason, it’s where we lived. In the end it was a very wise choice. It’s not a big city, but it’s not too small. The people don’t move around very much, which is essential if you’re doing a cohort study, if you have a lot of out migration, your sample is lost within a few years.
Which is how you’ve managed to keep a large amount of people involved in the study?
Exactly. It’s a good-sized city. It had around 230,000 inhabitants when we started, now about 400,000, which is still a reasonable size. We’ve always had very high response rates. The people we can’t interview are the ones who moved far away. If they live nearby they still come back to do the interviews. That’s why the last time we saw them, when they were 25 years old, over 4,500 out of 6,000 participated. It’s really amazing.
Why is the response rate so high?
The participants are really proud to be involved. We try to do quite a lot of publicity around the study and dissemination of the results. We get a lot of press coverage and there are always pieces in the newspaper and on the local TV. Earlier this year the biggest scientific programme on Brazilian television featured our cohort.
The poverty gap in Brazil as a country is enormous. How is it in Pelotas as a city?
Pelotas is pretty close to the national average, it’s a poor city in a rich area of Brazil. There is a huge gap between the rich and poor, although the gap is getting smaller. But still, when we did the study in 1982, the infant mortality among the poorest group of the population was seven times higher than the richest. Within the city you had these very distinct sub-populations – the differences were quite shocking. The first book we wrote about the cohorts in 1987 was called Epidemiology of Inequality because this was the most striking thing we came across when we analysed our data on child health.
Brazil is getting richer and with the World Cup and the Olympics being hosted there in the next few years there will be a huge spotlight on the country. How has it changed since 1982? Where will it be in 2016?
When we did the first cohort 20 per cent of the population were not entitled to any healthcare, other than in a charity hospital run by Catholic nuns. We now have a universal healthcare system, which we didn’t have in 1982. It still has problems and it’s not perfect, but at least everybody has access to healthcare. Brazil used to be number one in the world in terms of income disparities, famous for the gap between rich and poor, but it’s getting much better. We’re number 16 now. Still too high, but the trend is in the right direction, which is good to see.
You said there was a seven times gap in infant mortality between rich and poor. Is that getting closer?
Let me give you an example. In the poorest section of the population we had an infant mortality rate of 90 deaths per 1000 live births, now it’s just over 20. Infant mortality rates are dropping for the rich too, so the gap persists, but in absolute numbers the mortality has dropped much faster among the poor than the rich.
So the baseline is going up?
Yes. It’s very good news. The country getting richer has also affected child nutrition, which is much better now. But whenever you have a change like this, you have side effects. Now we have fat children…
Now you’re also involved in a comparison of cohort studies from low and middle-income countries.
In cohort studies it’s important to replicate the results. We applied to the Wellcome Trust for a grant, and I looked around the literature for five largest cohorts in low- and middle-income countries. By coincidence they are in very different parts of the world: Brazil, Soweto, New Delhi, Cebu [the Philippines] and Guatemala.
So I said, “Let’s do the analysis together”. I invited the other PIs and they all loved the idea. We then set up the COHORTS (Consortium of Health Orientated Research in Transitioning Societies) collaboration, with support from the Trust.
And your biggest finding in COHORTS was about rapid weight gain in the first two years of life.
We all got very consistent findings, mainly that these populations had low birthweights and were undernourished early in life. What we really found was that if babies put on weight and height in their first years of life it’s mostly beneficial. They have bigger brains, higher IQ, do better in school, they are taller as adults, they are more productive, they make more money, lots of good things on what we call ‘human capital’.
At that age range, most of the weight and height gain is very positive. But if they tend to put on weight after two years of life – and mainly after four to five years – it just builds up as fat mass. As a result they’re more likely to have diabetes, hypertension, obesity and so forth.
It’s quite interesting as you still see in a lot of developing countries that doctors are worried about children putting on weight as they’re worried about malnutrition. We showed that there is a window of opportunity – the first 1000 days of life.
Incidentally, do you know who made up the term? One of Hillary Clinton’s press officers! It’s a beautiful expression, which says that you have 270 days of pregnancy and 730 days [two years] to act to prevent undernutrition. This is your 1000 days, your window of opportunity. We’re really trying to get this message across to policymakers through the media and through the population in general. It’s very important. If you miss that window, there is irreversible damage that may be done.
Are you seeing this message get across?
I think the 1000 days idea is getting more visibility, but still we see paediatricians and nurses in low- and middle-income countries who see a child at three or four and think they should put on weight. The problem is that forcing a child to put on weight at this age has negative effects.
The window has closed?
The window has closed. That’s what we’re trying to disseminate now, a number of recent papers that have come out of the COHORTS collaboration are just repeating this message. We have to do a better job of disseminating, so I’m happy you’re picking this up. It’s important to reach outside of the medical journal readers.
Is it more difficult to carry out cohort studies in Brazil than in high-income countries?
Rich countries are much better organised and people can be followed up passively by the investigators. A good example is the NHS number. You can find out when the patient dies, or gets sick. We have to do everything by visiting them. The funding we received from the Wellcome Trust served as motivation to the Brazilian government to put some money into our cohort study. This has provided new buildings for our unit, just for the fieldwork.
So people come in to see you now, rather than you having to go and see them?
Yes. The buildings are funded the government, and all of the fieldwork is funded by the Trust.
You now have three cohort studies in Pelotas: ’82, ’93 and ‘04. What differences are you seeing between them?
Big things in obesity. Behavioural problems too; the kids are more aggressive. A lot of interesting findings are just starting to come out.
So you’re seeing clear changes between your groups?
We are. But many things are getting better: they participants are taller and healthier overall, their IQ’s gone up, you name it. But they are fatter and this will catch up with them later.
What about you Cesar? What gets you out of bed in the mornings?
I’ve always thought that action in prevention is better than treatment, although you need both, it’s not a simple dichotomy. My inclination was to always be more of a researcher than a clinician and designing strategies for prevention. To prevent you have to know how things are caused. You need to know about the 1000 day window if you want to improve adult health. I think in many ways, we started from nothing, just me and a colleague. Now we have this huge set of buildings and labs, and over 200 people work with us, so it’s great to have created something.
Of the work you’ve accomplished so far, what are you most proud of?
I think the fact that we got to do good work and managed to find the funding for it and that we got to become recognised and have publications in the Lancet, that sort of thing. From a scientific standpoint they’re pretty rewarding. From a personal aspect, it’s that I’m doing something relevant, the fact that we’re influencing policy – 130 countries in the world, now use growth curves that we made in a World Health Organization multicentre study that took place in six countries, including Pelotas. Now, if I walk into a clinic in London or New York, I can see our work.
I’m also President of the International Epidemiology Association and in the other half of my life, I work with the coalition ‘Countdown to 2015’, to plot how countries with high child mortality and high undernutrition are progressing towards the UN Millennium Development Goals. We’re taking the best available data, analysing them, and presenting them in a way that’s very simple to understand. Policymakers and people in health ministries can look at this and say, “This is what’s happening in my country”.
This is starting to look a lot like politics. Is that something you’ve ever considered moving into?
No, never. I love to do politically relevant scientific work, but I could never be a minister for health. I’ve been approached for this type of position, but no, this is not my thing. I think you should keep doing the things you do well.